Regressions Refresh

How do you calculate a residual?

• Actual - Predicted

• Predicted - Actual

Answer explanation

R.A.P

Residual = Actual - Predicted

I predicted someone would lose 14 lbs, but they actually lost 12 lbs. Calculate the residual.

• 14 - 12 = 2

• 12 - 14 = -2

• 14 - 2 = 12

• 2 - 14 = -12

I predicted someone would score a 78% on their test, but they actually scored an 81%. Calculate the residual.

• 81 - 78 = 3

• 78 - 81 = -3

• 81 - 3 = 78

• 3 - 81 = -78

A residual plot compares the residual values (dependent variable) to the predicted values. Which of the following residual plots indicate that the regression model is a "good fit" to the data?

Select two residual plots.

• 
• 
• 
• 
• 

Answer explanation

The residual plot of a linear scatter plot will have RANDOM scatter.

They do NOT have any patterns or clusters of points

The regression equation y = 25x + 100 models the cost, y, of an event for x people.

The table shows what the party sizes actually paid. Complete the table with the predicted values and residuals.

e

f

g

h

d

c

overestimate

35

15

underestimate

-35

-15

300

225

The r² value of the quadratic regression of this table is ...

• r²=0.9789

• r²=0.9515

• r²=0.9754

• r²=0.9613

Tags

R^2

Calculate the correlation coefficient (round it correctly to THREE decimal places!), then interpret its meaning

• r = -0.959

• r = -0.958

• r = 0.919

• As the number of staff increases, the average time to fill orders decreases

• As the number of staff increases, the average time to fill orders increases

Tags

Correlation Coefficient (r)

• strong positive

• strong negative

• weak positive

• weak negative

• This is an example of causation

An experiment was conducted to examine the relationship between two variables. The correlation coefficient r was determined to be -0.9542.

Which TWO statements best describes the relationship between the two variables in the experiment?

• There is a weak negative correlation between the two variables

• There is a strong negative correlation between the two variables

• One can conclude a change in the independent variable causes the change in the dependent variable

• One can not conclude a change in the independent variable causes the change in the dependent variable

What is most likely the correlation coefficient for this data set?

• r = -0.20

• r = -0.75

• r = 0.85

• r = 1.00